En-452414-5 ventricular conduction is a marker for arrhythmic risk in overlap sodium channel disease

SCN5A-E1784K (c.5350G>A) is the most common variant associated with LQT3 long QT syndrome (LQTS) and Brugada syndrome (BrS). It exhibits incomplete penetrance and can manifest variably as a mixed clinical phenotype of LQTS, BrS and/or conduction disorders. This presents a challenge for risk stratification and optimal management. The objective was to describe the clinical characteristics and risk markers in a large, well characterised international cohort of SCN5A-E1784K subjects and identify parameters that associate with event-free survival. Comprehensive clinical data, including initial presentation and follow-up data, were collected from SCN5A-E1784K subjects. Outcome ‘events’ included sudden death, aborted cardiac arrest, and documented VF, VT or arrhythmic syncope. ‘Lethal events’ included sudden death, aborted cardiac arrest or documented VF or VT. The effect of age at the time of ECG acquisition on ECG parameters was modelled and the residual, i.e. difference between observed value and the predicted age-corrected value, was calculated for each individual and referred to as residual PR, residual QRS, residual QTc and residual RR. Common variants previously associated with QRS duration were genotyped and a polygenic risk score (PRSQRS) derived. Associations between clinical characteristics, PRSQRS and event free survival were investigated using Cox proportional hazard regression models. SCN5A-E1784K subjects (n=231) were recruited from 20 centres. Four (1.7%) had a mixed phenotype of BrS, LQT3 and cardiac conduction disease. A mixed phenotype of BrS and LQT3 without any evidence of conduction disease was seen in 18.1% and a mixed phenotype of BrS and conduction disease, and Long QT and conduction disease, were seen in 3.1% and 12.9% respectively. There was no phenotype in 9.9% (includes n=6 with missing data). Median follow-up was 24.9 years (15.2 - 43.3) and 45 (19%) experienced an event. Lethal events were observed in n=14 (6%). Longer QRS duration (corrected for age) and PRSQRS were both associated with shorter event-free intervals (Table 1 and Figure 1). QTc and BrS phenotype were not associated with risk. Ventricular myocardial conduction, as evidenced by QRS duration and its genomic associations, appears to play a role in the risk of arrhythmic events in patients with SCN5A-E1784K. This provides an important opportunity for the personalisation of risk stratification of these patients, whose management can otherwise be challenging.Tabled 1Hazard ratios estimated in SCN5A-E1784K using Cox regression with adjustment for ethnicityVariableEventsLethal eventsHR [95% CI]p-valueHR [95% CI]p-valueMale1.33 [0.71-2.48]0.3692.22 [0.77-6.44]0.142Residual PR (ms)1.01 [1.00-1.02]0.2261.02 [1.00-1.03]0.012Residual QRS (ms)1.04 [1.02-1.06]<0.001*1.05 [1.02-1.09]0.002*Abnormal QRS-axis1.48 [0.69-3.19]0.3160.89 [0.19-4.10]0.879Residual QTc (ms)1.00 [0.99-1.01]0.9691.00 [0.97-1.02]0.664Long QTc0.87 [0.38-2.00]0.7461.05 [0.28-3.90]0.943Residual RR (s)1.10 [0.18-6.80]0.9159.51 [0.37-242]0.173Spontaneous type 1 Brugada ECG1.60 [0.50-5.11]0.4262.25 [0.50-10]0.290HR: hazard ratio; 95% CI: 95% confidence interval; ∗p-value reaching Bonferroni corrected significance threshold Open table in a new tab